CAJ | 학술논문

提出了一种径向基( RBF)神经网络的畸变校正方法。该方法利用RBF网络的以任意精度逼近连续函数的能力，通过采用张正友平面标定法得到样本数据训练网络，最终建立网络模型。该方法算法简易，适合于对摄像机捕获的图像进行几何校正。实验研究表明，该方法大大提高了校正精度。
제출료일충경향기( RBF)신경망락적기변교정방법。해방법이용RBF망락적이임의정도핍근련속함수적능력，통과채용장정우평면표정법득도양본수거훈련망락，최종건립망락모형。해방법산법간역，괄합우대섭상궤포획적도상진행궤하교정。실험연구표명，해방법대대제고료교정정도。
Proposed a radiate basis function ( RBF) neural network of the calibration method which has the ability of arbitrary precision approximate continuous function. The sample data is obtained by Zhang Zhengyou plane calibration method to using for training network. Then network model was estab-lished. This method is suitable for the image using of cameras to geometric correction. The experimental results show that the neural networks distortion correction technique has higher correction precision.